Tobacco smoking is a major cause of cancers at many sites, particularly the aerodigestive tract. For many years, filter and reduced tar cigarettes have been available, but have not resulted in reduced incidences of smoking-related cancers, because of compensatory smoking behavior. Recently, newer versions of "reduced-risk" cigarettes have been marketed, but their relative carcinogenicities are unknown. Cancer arises through multistage carcinogenesis, which involves the accumulation of genetic damage (such as mutations), over years, until a cell converts to cancer. Current assays for potential carcinogenesis in humans all have certain deficiencies. Few models employ normal human cells and mutagenesis is generally not an endpoint. To better understand risk from potential carcinogens, new model systems that can evaluate such risk are needed. In preliminary studies, we showed that in monolayer co-culture of normal human oral epithelial cells (NOE's) with a lac I rat reporter cell line (BB cells - that contain a lambda-based shuttle vector for detecting mutagenesis), increased mutagenesis induced by benzo(a)pyrene (BaP), relative to BB cells alone. This increased mutagenesis is most likely due to activation of BaP by the NOE's. We hypothesize that co-culturing NOE's with lac I reporter cells will allow detection of mutagenesis that is causally linked to metabolic activation of carcinogens by the normal human oral cells. To validate the model, mutagenesis induced by the two tobacco carcinogens (BaP and nitrosonornicotine) will be investigated. In Aim 1 we will establish and characterize a co- culture model composed of NOE's co-cultured with BB reporter cells. In Aim 2, this model will be used: 1) to compare mutagenesis induced by tobacco smoke condensate (TSC) from conventional cigarettes and recently-introduced "low-risk" cigarettes and 2) to determine inter-individual variation in the metabolism of these TSC's. Since human cells from a variety of organ sites can be grown in culture, the model has potential to become a flexible broad-based system to evaluate carcinogenicity risk of tobacco products in a number of human organs. A third aim will be a pilot study on the levels of cytochrome P- 450's in NOE's with and without the presence of TSC's, using gene expression arrays and confirmation by RT-PCR and western blotting. As NOE's are the very cells that are destined to develop into oral cancers, the model fills a void with the potential of being an organ specific technique to quantify mutagenesis, a major step in carcinogenesis. Clearly, it is preferable to determine their potential risk prospectively, rather than retrospectively. [unreadable] [unreadable] [unreadable]